The Spark

How Might We

Turn problems into opportunity-framing questions.

Converts a problem statement into open-ended design questions, each beginning with "How might we...?" "How" implies solvable, "might" acknowledges possibility not certainty, "we" signals collaboration. Each question opens a different direction without locking in a solution.

Quick Facts
Duration~20 minutes
CategoryThe Spark
OriginProcter & Gamble, 1970s
Created byMin Basadur (P&G)
Used in pathways
The Spark
Try How Might We → Download canvas PDF ↓

When to use it

Use as a bridge between understanding a problem and generating solutions. Especially when stuck, when the problem feels overwhelming, or when early brainstorming produced only obvious ideas.

It's the gateway tool in The Spark — reframing before ideating.

How it works in The Studio

Here's how a session works with WAiDE:

1
State the problem clearly
WAiDE helps you distill it to one sentence.
2
Generate first HMW questions
WAiDE prompts different angles.
3
WAiDE challenges scope
"Too broad? Too narrow? What if we flip it?"
4
Reframe and expand
Push into unexpected territory.
5
Select top 3–5 questions
Which ones open the most interesting doors?
6
Rank by potential
Which HMW would you most want to solve?

Sample output

Here's what a How Might We session looks like in practice:

Example: Employee Onboarding
Problem: "New hires take 6 months to become productive."
  1. 1HMW reduce the time to first meaningful contribution?
  2. 2HMW make onboarding feel like a real project, not a checklist?
  3. 3HMW use existing employees' knowledge without burning them out?
  4. 4HMW make new hires productive on Day 1 by changing what "productive" means?
  5. 5HMW create onboarding that experienced employees wish they'd had?
  6. 6HMW eliminate onboarding entirely by redesigning the role?
Most Promising
"How might we make new hires productive on Day 1 by changing what 'productive' means?" — challenges the assumption that productivity requires months of learning, opening space for immediate contribution.

What you get

5–8 reframed questions each opening a distinct ideation direction. Best HMW questions are neither too broad nor too narrow.

Foundation

Originated at Procter & Gamble in the 1970s by Min Basadur. Adopted by IDEO and the Stanford d.school. Used by Google, UK Government Digital Service, IBM, and INSEAD.

Procter & GambleIDEOStanford d.schoolGoogleIBMINSEAD

Why it works

How Might We questions work by doing something subtle but important: they hold the problem and the solution in the same sentence, at the same time. "How might we…" acknowledges that a problem exists and simultaneously declares that solutions are possible. This grammatical structure turns out to matter more than it sounds.

The alternative framings all fail in different directions. "We need to…" skips problem analysis and jumps straight to a predetermined solution. "Why can't we…" frames the situation as a constraint, triggering defensiveness. "How can we…" implies there's definitely a way, which can pressure teams into the first workable solution rather than the best one. "How might we…" keeps everything open — possibility without obligation.

The technique originated at Procter & Gamble in the 1970s, was developed further by IDEO, and became a core element of design thinking and Google Ventures' Design Sprint methodology. It's most powerful as a bridge: used after deep problem research (Empathy Map, Jobs to Be Done) to reframe insights as invitations, and before ideation tools (Crazy 8s, SCAMPER) to give those sessions a specific focus.

The calibration test: a good HMW question should generate at least 10 meaningfully different answers. If you can only think of one or two responses, the question is too narrow — you've already decided on the solution. If you have no idea where to start, it's too broad. Adjust until it's generative.

Frequently asked questions

Where did How Might We come from?

HMW questions were developed at Procter & Gamble in the 1970s by Min Basadur, then popularised by IDEO and the Stanford d.school as a core design thinking technique. Google Ventures embedded them in their Design Sprint methodology, which brought them to a much wider audience of product teams and startups.

What makes a good HMW question?

The right HMW is specific enough to give direction but open enough to allow genuinely different solutions. If you can only think of one answer, the question is too narrow — you've already decided on the solution. If you can't think of where to start, it's too broad. A useful calibration: a good HMW should generate at least 10 meaningfully different ideas.

How many HMW questions should I generate before choosing one?

Aim for 10–15 before narrowing down. Quantity first, then vote. The goal is a wide range of reframes — some will be too broad, some too narrow, some surprising. WAiDE will push you to keep generating when you want to stop early, because the most interesting HMWs are usually not the first ones.

What do I do after choosing a HMW question?

Move directly into ideation. HMW works as a launch pad for Crazy 8s (generate 8 ideas in 8 minutes), SCAMPER (apply structured creative lenses), or Worst Possible Idea (generate bad ideas and invert them). The HMW is not the answer — it's the permission to explore freely without committing to a direction.

Try How Might We?

WAiDE will guide you. About 20 minutes.

← Back to toolbox Start How Might We →